Figure 1.
Engraftment of human L363- and human RPMI8226- cells in vivo determined via flow cytometry.
A total of 106 mice received human L363 cells and 49 mice received human RPMI8226 cells via intratibial injection (it, L363: n=36, RPMI8226: n=15), intravenous injection (iv, L363: n=34, RPMI8226: n=10) or subcutaneous long bone implants (bi, L363: n=36, RPMI8226: n=24). Analyses were performed 35 days after tumor cell injection or when mice showed bad overall condition. Lack of NK cell activity, either by use of NSG mice or by use of mCD122-Ab treatment in NOD/SCID mice, and direct contact of implanted tumor cells with the murine bone marrow (BM), after it- or bi-injection, enhanced the engraftment of human MM cells. Human (h)HLA-A,B,C- as compared to hCD138-expression showed higher human tumor cell engraftment. This engraftment was most substantial in the BM, whereas peripheral blood (PB) and spleen engraftment was lower, albeit with similar patterns as in the BM. A. Mean tumor infiltration rates of human L363- cells in vivo determined via flow cytometry. In the BM, high human engraftment was obtained in NSG mice, irrespective of the injection modus. This could only be obtained in NOD/SCID mice, if pretreatment with mCD122-Ab and bi-implantation were performed. The PB engraftment after iv-injection of human L363 cells in NOD/SCID mice was most likely induced due to the plasma cell leukemia nature of these cells [28,29]. B. Mean tumor infiltration rates of human RPMI8226- cells in vivo determined via flow cytometry. Human RPMI8226 cells displayed similar engraftment patterns compared to L363 cells, albeit to a lower extend and highest BM engraftment potential with bi-injection regardless of the mouse strain. Differences between various conditions did not reach significance for all subgroups due to restricted group sizes.
Figure 2.
A. Detection of human L363 and RPMI8226 cells in immunodeficient mice using hCD138 antibody labelled with Alexa750.
20 mice per cell line were engrafted with L363 or RPMI8226 cells via different application routes. Images were taken once weekly for five weeks after implantation of the respective MM cell line with the Kodak Image Station in vivoFX. In addition to the determination of tumor load via CCD camera, animals were x-rayed and the two pictures merged for optimal localization of the fluorescent region. No x-ray was performed for the animal bearing RPMI8226 in this figure. Intraosseal tumor growth was clearly detectable by the in vivo imaging (IVI) system (red circles). Additionally, BM metastases became apparent within the adjacent tibia and lumbar spine (yellow circles). B. Detection of human L363 and RPMI8226 cells in immunodeficient mice using flow cytometry. IVI data for L363 and RPMI8226 were confirmed by flow cytometry 35 days after tumor cell injection showing 42% and 22% human HLA-ABC and CD138 positive cells in L363 and 21% and 20% for RPMI8226, respectively. Analyses were performed on metastatic BM lesions. C. Detection of human L363 and RPMI8226 cells in immunodeficient mice using immunohistochemistry. IVI data for L363 and RPMI 8226 cells were confirmed by immunohistochemistry specific for human CD138+ cells (CD138+-infiltrates, red arrows) Analyses were performed on metastatic BM lesions.
Figure 3.
A. Detection of human L363 and RPMI8226 cells in immunodeficient mice using hCD138 antibody labelled with Alexa750.
20 mice per cell line were engrafted with L363 or RPMI8226 via two different application routes (it and iv). Images were taken once weekly for five weeks after implantation of the respective MM cell line with the Kodak Image Station in vivoFX. Absolute net intensity (= pixel number within the region of interest) was markedly higher after L363 than RPMI8226 injection irrespective of the route of injection and genetic background of the recipient mouse (p<0.001). Moreover, NSG permitted higher engraftment than NOD/SCID irrespective of the injected cell line (p<0.002). B. Survival of NSG vs. NOD/SCID mice implanted with either L363 or RPMI8226 cells. 20 mice per cell line were engrafted with L363 or RPMI8226 via two different application routes (it or iv) in NOD/SCID or NSG mice (= 40 mice in total). Survival times of tumor bearing animals were monitored as an indicator of tumor burden. Survival of NSG compared to NOD/SCID after L363 or RPMI8226 injection was substantially shorter: the impact of the genetic background of the host was statistically significant for L363 (p<0.0001, Log-rank [Mantel-Cox]-test) as well as for RPMI8226 (p<0.0063*, Log-rank [Mantel-Cox]-test).
Figure 4.
Engraftment of MM patient vs. healthy donor-derived BM cells in NSG and in comparison to mock-injected controls determined by fluorescence-based IVI and flow cytometry.
A. Individual IVI images of NSG mice receiving MM patient or healthy donor-derived BM cells in comparison to mock-injected controls. As it-injection into NSG induced the most efficacious human cell engraftment, this approach was used for the implantation of MM patient-derived BM cells. IVI was performed using hHLA-A,B,C (day 38), hCD45 (day 46) and hCD138 Abs (day 56) labelled with Alexa750 dye. In addition, animals were x-rayed and the two pictures merged for optimal localization of the fluorescent region. Differences in BM engraftment capacities between mock-injected mice, healthy donor BM or MM-patient-derived BM cells were substantial: use of MM patient-derived BM cells induced sizeable engraftment, both at direct it-injection and to remote sites (upper panel in A). With use of healthy donor BM cells, hCD45-and hHLA-A,B,C-, but not hCD138-Ab-IVI, also revealed minimal engraftment. Therefore, hCD138-Ab-IVI revealed most substantial differences between MM, healthy donor and mock-injected cells, showing engraftment with use of MM patient BM cells, but no engraftment with healthy donor or mock-injected cells. B. Mean net intensity values determined by fluorescence-based IVI of NSG mice receiving MM patient or healthy donor-derived BM cells in comparison to mock-injected controls. IVI was performed using hHLA-A,B,C (day 38), hCD45 (day 46) and hCD138 Abs (day 56) labelled with Alexa 750 dye. Differences in BM engraftment capacities between mock-injected mice (light grey bars), healthy donor BM (green bars) or MM patient-derived BM-cells (blue bars) were significant (p<0.001). In mock-injected mice, no fluorescence was detected. Healthy donor BM cells could be detected exclusively in the BM, showing no extramedullary engraftment over the entire observation period: After MM patient-derived BM injection, IVI using hHLA-A,B,C- and hCD45-Ab indicated substantial human engraftment as compared to the hCD138-Ab. Flow cytometry helped to confirm the results as shown for a representative patient in A, revealing that CD138 expression on d56 was substantial and significantly higher as compared to mock-injected and healthy donor mice (C). C. Mean infiltration rates determined by flow cytometry of NSG mice receiving MM patient or healthy donor-derived BM cells in comparison to mock-injected controls. IVI data from each individual animal was verified by flow cytometry: 56 days after MM patient-derived BM cell injection, BM, PB and spleen specimens were analysed. Human engraftment was defined as >0.08% hHLA-A,B,C positivity (= twice the maximum percentage of false-positive cells in mock-injected mice [0.04%]). Injection of BM cells from MM patients resulted in substantially more hHLA-A,B,C, hCD45 and hCD138 positive cells than that of healthy donor BM which was significant for hCD138+ cells in the BM, and all three tested markers in the spleen. Furthermore, all investigated markers were significantly higher expressed in BM and spleen of animals bearing MM patient-derived cells compared to mock-injected animals (Student´s t-test, p<0.05). Mean infiltration rates of hCD138+ cells in BM, PB and spleen sites were 1.02%, 0.25% and 0.08%, respectively, which appeared fairly substantial, and which were substantially larger than that after healthy donor BM use. Healthy donor-derived BM cells induced very minor BM engraftment (mean infiltration rate of hCD138+ cells: 0.39 ± 0.45).
Figure 5.
A. Chemosensitivity of L363 cells in NSG against different antimyeloma agents determined by flow cytometry.
A total of 24 NSG received it-injections of 2x105 L363 cells: seven days thereafter, 18 mice were randomized into three groups of six animals each. Six additional mice were analysed on the first treatment day (d7) to determine tumor load before the respective therapies. MM cell engraftment was determined by flow cytometry using hCD138-Ab (n = 2-3 per group and time point, 4 measurements per group over the time). Black: mice were treated with control vehicle (0.9% NaCl, i.p. d7-11, 14-18). Yellow: mice were treated with 3 mg/kg/d dexamethasone, i.p. (d7-11, 14-18). Green: mice were treated with 0.7 mg/kg/d bortezomib, iv (d7, 11, 18). 7, 14, 28 and 35 days after implantation, tumor cells were assessed via flow cytometry and increased detectably. Analyses on day 35 confirmed that tumor growth was reduced with dexamethasone by 51% and with bortezomib by 66%. B. Chemosensitivity of L363 cells in NSG against different antimyeloma agents determined by fluorescence-based-IVI. A total of 15 NSG received it-injections of 2x105 L363 cells: seven days thereafter, mice were randomized into two groups of six and nine animals, respectively. IVI using Alexa750 tagged hCD138 antibody was employed to determine tumor engraftment and treatment efficacy. In addition animals were x-rayed and the two pictures merged for optimal localization of the fluorescent region. Tumor growth after bortezomib treatment on day 18 was significantly reduced which was observed both for primary and metastatic sites (Student`s t-test, one-tailed p< 0.05). Error bars depict standard deviation.
Figure 6.
Chemosensitivity of patient-derived MM cells in NSG against different antimyeloma agents determined by fluorescence-based-IVI (A-B) and flow cytometry (C) compared to clinical outcome of the donor patient (D).
A total of 30 NSG received it-injections of 2x106 patient derived MM cells: 11 days thereafter, mice were randomized into three groups of ten animals each. Black: mice were treated with control vehicle (0.9% NaCl, i.p. d11-15, 17-22). Yellow: mice were treated with 3 mg/kg/d dexamethasone, i.p. (d11-15, 17-22). Green: mice were treated with 0.7 mg/kg/d bortezomib, iv (d11, 15, 19, 22). A. Tumor load determination using fluorescence based IVI. IVI using Alexa750 tagged hCD138 antibody was employed to determine tumor engraftment and treatment efficacy from day 11 (first day of treatment) on once weekly. In addition animals were x-rayed and the two pictures merged for optimal localization of the fluorescent region. Non-tumor bearing animals were treated accordingly and used as negative control. Error bars depict standard deviation. Determination of net intensity over time revealed a reliable engraftment of patient-derived MM cells over time as compared to mock-injected mice. Treatment with bortezomib or dexamethasone reduced tumor progression compared to untreated control. Tumor growth after bortezomib and dexamethasone treatment on day 26 was significantly reduced by 40% and 22%, respectively, which was observed both for primary and metastatic sites (one way ANOVA, p<0.0001). B. MM cell engraftment in the BM determined by flow cytometry. MM cell engraftment in the BM was determined by flow cytometry using hCD138-Ab on the last experiment day (d=38, n=5 per group). Analyses confirmed that tumor growth was reduced with dexamethasone and bortezomib. C. Immunhistochemical stainings of a femur from a NSG mouse inoculated with MM patient-derived BM cells. Formalin fixed and paraffin embedded long bones from tumor bearing mice were stained with HE, anti-human CD38-, CD138- and kappa-Ab. The patient's kappa/CD38/CD138 triple-positive BM cells were well detectable in NSG mice 35 days after tumor cell inoculation. IHC analysis for hCD138 expression confirmed IVI and flow cytometry results as depicted in 6A+B. D. Specific response parameters in the patient after treatment. Specific response parameters in the patient after treatment. improved after three cycles of bortezomib, cyclophosphamide and dexamethasone (VCD) induction. A decline in his total protein (TP), gamma-globuline fraction (g-Gl), IgG, BM plasma cells (BM-PCs) and kappa-light chains (serum free light chains, [k-SFLC]) was induced as determinants of VCD response. Results are given as fold decreases from baseline levels.